Network Re-construction For The Complex Data Generated From The Discrete And Continuous Models

Network Re-construction For The Complex Data Generated From The Discrete And Continuous Models

Network Inference for complex systems is crucial to infer connectivity among variables in many subject areas, ranging from finance to health sciences. Therefore, it is a rapidly developing area with newly proposed methods.

In this seminar talk, Huseyin Yildirim will present the Mutual Information (MI), double normalised Mutual Information Rate (MIR) methods and their lagged versions to reconstruct the initial network for artificial data generated by the coupled logistic map, coupled circle map, and coupled Hindmarsh-Rose (HR) model of neuronal activity.

The authors in [1] have already showed that the double normalised MIR can capture all links in the original network for discrete and continuous dynamical models when specific conditions are met. Our study proposes that the lagged versions of MI and double normalised MIR can infer network topology 100% successfully for small time series.

Finally, our results show that the latter methods have better performance when using the instantaneous frequency of the membrane potential in the HR model as a probe to infer network structure.

Speaker

Huseyin Yildirim, University of Essex

How to attend

If not a member of the Dept. Mathematical Science at the University of Essex, you can register your interest in attending the seminar and request the Zoom’s meeting password by emailing Dr Jesus Martinez-Garcia (jesus.martinez-garcia@essex.ac.uk)